ongonite and topazite dikes in the flying w ranch area ...see fig. l). beryllium mineralization, in...

American Mineralogist, Volume 73, pages 507-523, 1988

Ongonite and topazite dikes in the Flying W ranch area, Tonto basin, Arizona

WrNrrnro T. Konrntrnrnn, DoNlr.o M. Bunr*Department of Geology, Arizona State University, Tempe, Arizona 85287, U.S.A.

Ansrnacr

Topaz-rich ongonite and topazite dikes cut rocks ofthe Proterozoic Alder Group in theFlying W ranch area, 12 km west of Young, Arizona. The main ongonite dike, composeddominantly of quartz, topaz, and sodic plagioclase, is mineralogically similar to ongonitesdescribed from Mongolia and the Transbaikal region of the USSR. The topazite dikes,composed predominantly of quartz and topaz, are mineralogically and compositionallysimilar to topazite from the New England district, New South Wales, Australia. The twotypes ofdikes in the Flying W ranch area grade into each other over a few meters. This isthe first reported occurrence of either of these rock types in North America, and their firstreported occurrence together.

Field, textural, mineralogical, and geochemical evidence suggests a dominantly mag-matic genesis for both ongonite and topazite dikes. This evidence includes crosscuttingrelations with Alder Group rocks, chilled and flow-foliated dike margins, xenoliths ofAlder Group rocks along dike maryins, and dike topaz with an igneous texture and geo-chemical signature.

The melts that formed these dikes are believed to be extreme differentiates of an F-richgranitic magma. The paucity of aplitic and pegmatitic textures in the dike rocks indicatesthat the melts were relatively "dry" (not accompanied by an immiscible fluid phase) upontheir emplacement as dikes. Crystallization was, however, accompanied by the exsolutionofa late fluid phase that altered surrounding Alder Group rocks. These fluids, enriched inBe, F, B, K, and Na, formed two areas of significant alteration and mineralization referredto as the Breadpan and Dysart areas. The discovery ofacicular blue beryl in zones alteredto tourmaline, quarlz, alkali feldspar, and fluorite prompted exploration of both areas inthe early 1960s.

Topaz was discovered in a unit of the Alder Group in the 1950s. The topaz occurs withq\artz, sericite, and fluorite as microveinlets cutting metamorphosed rhyolite tuf andnearby rocks. It is considered to be a product ofthe mineralizing event above, rather thana primary, vapor-phase mineral.

Dike emplacement and associated mineralization appears to have occurred after theMazatzal orogeny (1715 Ma) and before the deposition ofApache Group rocks (l 100 Ma).

fNrnooucrroN

In the Flying W ranch area, Proterozoic Alder Grouprocks host a series oftopaz-rich dikes and associated Be-Fmineralization. The area is located l2 km west of Youngand 26 km southeast of Payson, Arizona, in the Tontobasin, a part of Arizona's Transition zone (Pierce, 1985;see Fig. l).

Beryllium mineralization, in the form of blue, acicularberyl, was discovered at two localities in the Flying Wranch area (Breadpan and Dysart areas, see Fig. 2) in theearly 1960s (Smith, 1962; Meeves, 1966). At a third lo-cation, here referred to as the Spring Creek locality (seeFig. 2), L. T. Silver discovered minute grains of topaz

* Visiting Scientist, Lunar and Planetary Institute, 3303 NASARoad 1, Houston, Texas 77058, U.S.A.

0003--004x/88/05064507$02.00

(AlrSiOo(F,OH)r) in heary-mineral separates from ametarhyolite unit (Flying W rhyolite) in the Alder Group(cited in Conway, 1976). The occurrence of topaz inseemingly unaltered rhyolite and the proximity of thisrhyolite to Be-F mineralization led Burt et al. (1982) tosuggest that the Flying W rhyolite might be a Precambri-an analogue to Tertiary topaz rhyolites of the westernUnited States and Mexico (Christiansen et al., 1983,1986). A recognized problem with this proposal was thefact that topaz in the Flying W Formation occurs in ametamorphosed rhyolitic tuff, whereas topaz in Tertiarytopaz rhyolites is nearly always found in lava.

This study was initiated to determine the origin of thetopaz in the Flying W rhyolite. The discovery of topaz'rich dikes in both the Dysart and Breadpan areas prompt-ed expansion ofthis study to include the petrogenesis ofthe dikes and their relation to mineralization.

507

€^ I -"/^

ooto pro!,ix Vo,i*

508 KORTEMEIER AND BURT: ONGONITE AND TOPAZITE DIKES

35"

34o

Fig. |. Location map of study area showing Arizona's phys-iographic provinces (redrawn from Pierce, 1985).

Rocr uNrrs

Alder Group

Both the Be-F mineralization and topaz-rich dikes inthe Flying W ranch area are contained within the twolowermost members of the Alder Group, the Breadpanand Flying W Formations. These two formations consistof metamorphosed volcanic and clastic sedimentary rocksthat exhibit near-vertical dips and are, as a rule, well-foliated. Metamorphism and deformation of these rockstook place during the Mazatzal orogeny (Wilson, 1939),between 1660 and 1715 Ma (Silver, 1964).

Rocks of the Alder Group were first mapped, and for-mations and units first named, by R. G. Gastil (1954).More detailed descriptions of these rocks can be found inGastil (1958), Conway (1976), and Sherlock (1986).

Breadpan Formation. The Breadpan Formation, namedby Conway (1976) for extensive exposures in BreadpanCanyon and previously referred to as the Alder Forma-tion (Gastil, 1954), is composed entirely of meta<lasticrocks finer than cobble conglomerate (Conway, 1976). InBreadpan Canyon and vicinity (see Fig. 2), the formationconsists predominantly of purple to gray phyllite (com-posed largely of sericite), interbedded with metagray-wacke and meta-pebble conglomerate. Small (0.5 to 3 mwide), discontinuous, metamorphosed pegmatitic lensescomposed of quartz and K-feldspar are also common.These lithologic units (and to a lesser extent the pegrnatitelenses) exhibit a well-developed, steeply dipping foliationtrending approximately N60'E across the map area.

Nonfoliated, unmineralized quartz veins, ranging inwidth from less than a few centimeters to greater than 4m, occur throughout the Breadpan Formation, but do notappear to be directly related to the Be or F mineraliza-tion.

Thin-section examination of phyllite from the Bread-pan Formation shows that the degree of metamorphismof these rocks corresponds to the chlorite zone of thegreenschist facies (Jackson, 1970). No minerals that wouldindicate a higher grade of regional metamorphism havebeen found in any Alder Group rocks, nor are there anyvisible signs of contact metamorphism to a hornfels by aburied pluton.

Flying W Formation. The Flying W Formation, alsohost to the beryllium-fluorine mineralization in the area,conformably ovedies the Breadpan Formation. It rangesin thickness from 100 to 400 m and consists ofinterbed-ded basaltic lavas and pyroclastic rocks, rhyolitic tuflandesitic lava, and coarse conglomerates. Deformation-produced foliation is either weak or absent in these units.A measured columnar section (Fig. 3) was taken fromoutcrops on the west side of Spring Creek, I km north-northwest of the Flying W ranch house (see Fig.2).

The lower to middle third (70-100 m) of the Flying WFormation (Fig. 3) consists of the Flying W rhyolite, amassive tuffaceous unit. The apparent age (U-Pb zircon)of this unit (L. T. Silver, cited in Conway, 1976) isl7l5(15) Ma. This is the only date available for rocks inthe study area. The heavy-mineral separates used in thisdating process also yielded abundant, minute topaz. Theoccurrence of topaz was confirmed by R. G. Gastil (pers.comm., 1985), who discovered (but did not report) topazin heavy-mineral separates from the same unit in the mid-1950s (see Gastil, 1954, 1958). Topaz was not identifiedin hand specimen nor in this section, and the origin ofthe topaz remained unknown.

Ongonite and topazite dikes

A series of aphanitic, topaz-ich, felsic dikes have beendiscovered in the map area near both the Breadpan andDysart beryllium claims (Kortemeier and Burt, 1985; seeFig. 2). These dikes are narrow (less than 5 m), steeplydipping bodies that are fairly continuous and well-ex-posed along their strike lengths of from 250 m to 2.4km.The interiors of the dikes are white, fine-grained, andgranular. This massive interior rock weathers to a sandygrus. Along the margins of the dikes, in a zone that isgenerally 2-8 cm wide, the dike is pink, extremely fine-grained, and flow-banded (Fig. 4). This dense, very fine-grained zone exhibits conchoidal fracture. Locally, thedike interiors have a purple tint when freshly broken dueto disseminated fluorite. This color disappears on weath-enng.

Xenoliths of surrounding rocks are common in themarginal zones of the dikes (see Fig. 8), and the contactsof the dikes with the xenoliths and with surroundingcountry rocks are sharp (Fig. 4). Visible contact meta-

320

509KORTEMEIER. AND BURT: ONGONITE AND TOPAZITE DIKES

s6xC 6Q J

5 t3 ;o o

! ;

o

o

E >. = 6> €: . b€ faaEEb 6; i iU O

P I 3 E

E E5I

,-vF

tbI)

6

N

h

(J

!?

I

BbI)

lJro

Iu

o

6i

b0

s6 .

E g

\ r: . . v |' . . \ \t l\ \

5€g->? o ai

iE )€\ F

5 1 0 KORTEMEIER AND BURT: ONGONITE AND TOPAZITE DIKES

l 5 0 m0 l s

Fig. 3. Measured columnar section of the Flying W Forma-tion, Spring Creek locality.

morphism of the country rock either consists of minorsericitization in a zone less than 2 cm wide or is absent.

The dikes are texturally homogeneous; no differentiat-ed aplitic and pegmatitic phases are present. In thin sec-tion, the dikes are either fine-grained porphyritic oraphanitic, with grain sizes rarely exceeding I mm. Noevidence of postemplacement metamorphism is present;i.e., deformation-produced foliation is abseit, veining andfracturing of the rocks are minor, and, in thin section,quartz grains are unstrained. Vein-controlled hydrother-mal alteration also appears to be absent.

The extremely leucocratic dikes in the Flying W rancharea show a striking textural resemblance to ongonite dikesdescribed from Mongolia and the Transbaikal region ofthe USSR (Kovalenko eI al., 1971,1975; Kovalenko andKovalenko, 1976) and discussed below. Mineralogically,the dikes can be divided into two groups: those consistingpredominantly of quartz and topaz, with variable mica(referred to as topazite) and those consisting predomi-nantly of quartz, topaz, and albitic plagioclase, with vari-able mica (referred to as ongonite). These dikes are de-scribed individually below.

Fig. 4. Photograph of flow-foliated Breadpan dike maryin.Note dike's sharp contact with country rock. Width of dike shownis about 20 cm.

Topazite dikes of Breadpan Canyon. Two dikes havebeen discovered on the northwest slope ofBreadpan Can-yon. The first to be discovered, here called the Breadpandike, is also the most elongated; it crops out over a strikelength of approximately 2.4 km (Fig. 2). The Breadpandike is 1.5 to 2.0 m wide, trends approximately N45"E,and dips approximately 54'NW. This orientation is sub-parallel to the N60"E, 70"NW orientation ofthe Breadpanphyllite in the area. On an outcrop scale, however, theBreadpan dike clearly cuts across and warps the slaty fo-liation. Contact metamorphism of the phyllites by theBreadpan dike is limited to a zone of sericitization lessthan 2.5 cm wide.

In thin section (Fig. 5), the Breadpan dike consists pre-dominantly of two minerals, fine-grained (< I mm) an-hedral quartz (700/o) and finer-grained (<0.7 mm) blockyto prismatic topaz (30o/o). White mica is locally abundant(up to 150/o), and its texture appears to be primary. Fluo-rite, visible in hand specimen, is also present locally asindividual cubes or as granular aggregates. Opaque min-erals are extremely rare. Toward the western end of theBreadpan dike, the dike rock contains minor quantities

50

- =7=-= -- '- -'Conglomerate

2--"=3:'_-'4: o 4 . o . 4

O o ^ A 9 ' 4 ea e a a a - 3 +n g l o m e r a t e a - a -

=-s=:---;-;::- - -

KeratophvTE--, - ^

:b-o- o:- o :-- =_ 3 =\

oo*oo * i - -ol*-^- ' -

o o o o 4 o ' _ e o n o - 4 t - o - Q

9 -

C o n g l o m e r a t e e - - 1 ' o

Basal t ic LavaBrecc ia

KORTEMEIER AND BURT: ONGONITE AND TOPAZITE DIKES 5 l l

of unaltered plagioclase, either as phenocrysts in the fine-grained, quaftz-topaz groundmass or as grains intergrownwith equigranular quartz and topaz.

Major- and trace-element analyses of four samples ofthe Breadpan dike topazite (samples containing no feld-spar) are given in Table l. In confirmation of the petro-graphic observations, the rock contains unusually highSiO, (71-860/o) and AlrO. (8-2lo/o) and is relatively lack-ing in all other major elements. F, at or above 20lo (theupper detection limit was set at 20000 ppm by the ana-lyst), is high owing to the abundant topaz. The rock isenriched in Sr and Ta, and depleted in Rb, Be, Y, Nb,and Mo relative to peralkaline rhyolites, topaz rhyolites,and ongonites (see Christiansen et al., 1983).

A second dike, located approximately 100 m upslopefrom the Breadpan dike, is similar to the Breadpan dikein all but two respects: (1) the second dike is exposed ononly one hillside, providing approximately 250 m of hor-izontal exposure (compared luv.rth 2.4 km of exposure onthe Breadpan dike), and (2) weathered, limonite-replacedpyrite cubes occur locally in the border zone ofthe seconddike. No pyrite has been observed in the Breadpan dike.

These two dikes, the Breadpan dike and its less elon-

Fig. 5. Photomicrograph ofBreadpan dike, centralzone. Noteacicular, high-relief topaz. Remainder of rock is composed oflarger, anhedral quartz. Plane-polarized light. Field ofview about700 pm across.

gate neighbor, are geochemically and mineralogicallysimilar to rocks referred to as "topazite" by Eadingtonand Nashar (1978) and Eadington (1983). A completediscussion of topazites is given below.

TreLe 1. Maior- and trace-element analyses of rocks from the Breadpan and Dysart dikes, Arizona, of ongonite from Mongolia,

and of tooazite from New South Wales. Australia

Topazites

61 01 61 03 61 07 Ongonite

sio,Alr03FerO.FeOMnOMgoCaOLi,ONaroK"oTio,P,O.

79.7615.480 5 00 . 1 80.010.140.07

0.040.320.040.33

>2000032

2555648

159< lT 63

5 5

1 4< l541 61 223.2

0 .1381

0.33.4

86.938.650.400.180.010 . 1 10.07

o . 1 70.340.o20.21

1 65004586

<5034

179< 179

o731 0< 11 1 6321 81 2

71 0021.30o.470.12

<0.010.29u,zo

0.011.740.020.22

>200001 1 5138<50

47133< 1R I

2471 5< 1741 81 19.1

0.83134

0.354.1

79.0217.770.380.04

<0.010.03

<0.01

<0.010.070.020.22

>200001 2

1 1 3<50

2377.5< 122< 1291 3< 13 I

1 65.26.6

0 .11500.30J . O

75.5418.090.670.210.020.070.42

4.320.410.020.07

1 9000263152

1 200103

1 . 5< l

3272

1021 1< 140

I3 .1

1 1

1 . 714.10.944.5

70.7816.78o.270.260.180.200.340.425.243.31

0.07

1 99001 876

20za6967

19.237

6610.7

93.815 .11 .016.1

73.4020.800.000.560.020.000.00

8 1 . 1 015.80

0.400.01o.170.14

0.010 . 1 00.03

46000FRbSrBaNb

LiBeSn

MoZlHfUTh

<0.010.04o.02

65500

1 0

4130

650

1 5400

660

Rb/SrK/RbNb/TaZrlHt

0.52oo

0 .19J O

Note: samples 6101 , 6103, 6107, and 61 Og are Breadpan dike topazite; 61 01 and 61 09 are from the coarse, interior zone, 6103 and 6107 are fromthe marginal; banded zone Sample 693 is Dysart dike ongonite, marginal zone. Oxides, Be, Li, and Mo analyzed by the direct current plasma method;F by specific ion; Rb, Sr, Nb, Y, and Sn by xnr; Mo, Ba; Hf, Ta, W, Th, and U by neutron activation. All analyses are by Bondar-Clegg, Lakewood,Colorado; analytical error not specified. Ongonite is from the Ongon-Heierhan district, Mongolia (from Kovalenko and Kovalenko, 1976) and topazitesare from the New England district, Australia (from Eadington and Nashar, 1978). Oxides in wto/o; trace elements in ppm.


Fig. 6. Photomicrograph of elongate, radiating topaz grains,Dysart dike. Plane-polarized light. Field ofview about 700 pm

The Dysart ongonite dike. A third major dike was dis-covered approximately 2 km northwest of the Flying Wranch house and I km west of Spring Creek. This dike,here named the Dysart dike, intrudes conglomerates andbasalls of the Flying W Formation (Fig. 2). The dike is 4to 5 m wide and crops out over a strike length of ap-proximately 700 m. It has an orientation of NIO'W, 55'NEand clearly cuts across the N50'E, 80'NW foliation of theFlying W units. An 8- to l5-cm-wide dikelet, exposed forapproximately 6 m along strike and located 2l m east ofthe main dike, trends N35oW, or almost perpendicular tothe foliation ofthe country rock.

Approximately 300 m south of the northern end of theDysart dike, small pyrite cubes (less than 5 mm and par-tially replaced by limonite) are abundant in both margin-al and central dike material.

In thin section, the Dysart dike is composed of ap-proximately 400/o anhedral quartz, 30o/o Iopaz, and 300/oalbite. White mica and fluorite are present locally, buttogether constitute less than 50/o of the rock. The topazgrains, here up to 0.7 mm long, occur as blocky or pris-matic grains or as glomeroporphyritic, radiating crystalsdisplaying excellent basal cleavage (Fig. 6). The albitegrains are subhedral to anhedral and appear to be partlyresorbed. The upper northern end ofthe Dysart dike con-tains only quartz, topaz, and variable white mica; plagio-clase is noticeably absent. In a transition zone exposedon a steep slope between upper plagioclase-barren andlower plagioclase-bearing dike rock, relict plagioclasegrains are now altered to quartz, sericite, and variabletopaz (Fig. 7). In fluorite-rich areas in this transition zone,plagioclase has been altered to the assemblage fluorite,sericite, and quartz.

A geochemical analysis of a sample of Dysart dike on-gonite is given in Table I (sample 693). The Dysart dikeresembles the Breadpan dike in its high content of SiO,(7 5.5o/o) and AlrO,; the presence of abundant albite addsappreciable NarO to the analysis. F is again approxi-

Fig. 7. Photomicrographs of relict plagioclase grain, Dysartdike. Replacement ofplagioclase here is by quartz, sericite, andtopaz. Top, plane-polarized light. Bottom, crossed nicols. Fieldofview about 700 pm across.

mately 20lo owing to the high content of topaz. Comparedto peralkaline rhyolites, topaz rhyolites, and ongonites(see Christiansen et al., 1983), this rock has high concen-trations of Sr, Li, Sn, and Ta, and low Be, IJ, Y, Mo, andZr values. The Dysart dike is mineralogically and geo-chemically similar to ongonites described by Kovalenkoand Kovalenko (1976) and below.

Dikelet east of Spring Creek. One very narrow (<15cm), short (outcrop <15 m) dike has been discoveredcutting Breadpan slate in a small canyon on the east sideof Spring Creek, approximately 1.3 km northeast of theFlying W ranch house (see Fig. 2). Where sampled, thedikelet is 3.5 cm wide and flow-banded from margin tomargin. Small xenoliths of the surrounding slate are pres-ent in this dikelet and locally distort the flow banding(Fie. 8).

In thin section; this dikelet is porphyritic, with an ex-tremely fine-grained, flow-foliated groundmass consistingof approximately 25o/o topaz and 7 5o/o quartz (+ plagio-clase?). Euhedral to subhedral quartz, subhedral to an-hedral plagioclase, microperthitic alkali feldspar, andmuscovite occur as phenocrysts in this dikelet. This is theonly recognized occurrence ofpotassic feldspar in the dike

';F-

KORTEMEIER AND BURT: ONGONITE AND TOPAZITE DIKES 5 1 3

rocks of the Flying W ranch area. The presence of bothplagioclase and microperthitic feldspar makes this dikeletvery similar 1o Mongolian ongonites.

AITnru.IToN AND MINERALIZATION

Three areas of alteration and mineralization in theFlying W ranch area are discussed below. Two of these,the Breadpan and Dysart areas, were prospected for be-ryllium in the early 1960s. The third area, located alongSpring Creek and here referred to as the Spring CreekIocality (see Fig. 2), is not strongly mineralized in beryl-lium, but is the locality where topaz was discovered inheavy-mineral separates from the Flying W rhyolite(Conway, 1976). Weak alteration (fluorite or tourmalineor both with sericite along veinlets) occurs throughout thestudy area. The mineral-resource potential of this areaand the nearby Hells Gate Wilderness Area is discussedin conway er al. (1983).

Breadpan area

In Breadpan Canyon, beryllium mineralization is host-ed by Breadpan Formation phyllites. The mineralizationinvolves pervasive alteration and veining ofthe countryrock in a zone approximately 1.5 m to 12 m wide andextending for approximalely 2 km parallel to the foliationof the host rocks (see Fig. 2). Beryllium exploration inthis area included surface sampling, trenching, and drill-ing over the strike length of the mineralized zone. Theberyllium grade is referred to as "marginal," and no pro-duction is recorded (Smith, 1962).

The alteration assemblage in the Breadpan area is com-posed predominantly of black, fine-grained tourmalineand pink to white K-feldspar, with subordinate fluorite,quartz, chlorite, calcite, sericite, beryl, and specular he-matite. Alteration in the Breadpan area is weakest alongthe margins of the mineralized zone and intensifies to-ward the center. Along the margins, the Breadpan slatehas been sericitized and feldspathized, producing a light-colored, coarse-grained phyllite. Stronger alteration hascreated a finely banded rock composed of layers of thelight-colored, coarse-grained phyllite separated by tour-maline-rich layers that are oriented parallel to the originalfoliation. In the center of the mineralized area, the mostintense alteration has caused the original slate to be com-pletely replaced by alkali feldspars, with or without tour-maline and chlorite. This core area is also cut by veinletsof tourmaline, calcite, and fluorite. Beryl occurs as aggre-gates of small (<3 mm), blue grains intimately associatedwith pink alkali feldspar, black tourmaline, brown car-bonate, and clear quarlz.

The altered and mineralized zone is located approxi-mately 45 m downslope from the Breadpan dike, yet theintervening country rock is completely unaltered. Evi-dently the F, Be, and B-rich mineralizing fluids were lo-calized by fractures parallel to the foliation and arosefrom depth, rather than directly from the adjacent dike.

Fig. 8. Photograph ofdikelet east ofSpring Creek. Note xe-nolith-rich margin.

Dysart area

In the Dysart area (Fig. 2), beryllium mineralization ishosted by Flying W Formation conglomerates and ba-salts. The younger Houdon Formation quartzite, whichoccurs in this area in both fault and conformable contactwith the Flying W rocks (Fig. 2), does not appear to hostmineralization. Claims were staked in early 1960s by thesame group that conducted the exploratory trenching anddrilling in the Breadpan area (Smith, 1962). Four to fivepits were excavated, but no further exploration was at-tempted.

The country rock surrounding the Dysart dike is stronglyaltered, especially on the hanging-wall side ofthe dike tothe northeast. The alteration assemblage is dominated bytourmaline, fluorite, and quartz, with subordinate K-feld-spar, sericite, beryl, and topaz. The alteration appears tohave commenced with feldspathization, sericitization, andsilicification as in the Breadpan area. Thereafter, miner-alization in the two areas diverged. In the Dysart area,the second stage consisted of brecciation of the countryrock and deposition of tourmaline and fluorite (with orwithout quartz and sericite) as the supporting matrix forthe previously altered clasts (see Fig. 9). A third stageconsisted offracturing ofthe breccia, followed by veiningby tourmaline + fluorite + qvarlz + beryl. The berylmineralization represents the final pulse of fracture filling;beryl usually occurs as dark blue, elongate, radiating crys-


Fig. 9. Photograph of tourmaline- and fluorite-supportedbreccia, Dysart area. Clasts are composed ofsilicified and feld-spathized Flying W conglomerate. Small quartz vein at base ofrock.

tals up to 2.5 cm in length (Fig. l0) on fractures havingan approximate orientation of N20'E, 25'NE.

The mineralization in the Dysart area is more perva-sive than that in the Breadpan area. The Flying W unitsin the Dysart area are only weakly foliated, so that min-eralizing fluids tended not to be controlled by the orien-tation ofthe host rocks (in contrast to the Breadpan area)and, instead, brecciated the competent Flying W units.

Spring Creek locality

The outcrop of Flying W units located I km north-northwest of (downstream from) the Flying W ranchhouse, on the western bank of Spring Creek, is here re-ferred to as the Spring Creek locality (Fig. 2). The beryl-lium mineralizalion in this area is hosted by Flying Wrhyolite, conglomerate, and basaltic breccia units. Themineralization and alteration in this area, neither per-vasive nor intense, consists of small veins and coatedfractures locally surrounded by narrow, weakly alteredzones. The alteration assemblage includes quartz, tour-maline, sericite, topaz, fluorite, and beryl.

Macroscopic evidence of alteration-mineralization atthe Spring Creek location includes fluorite on fractures inthe Flying W rhyolite and overlying conglomerate, vein-lets of tourmaline, quartz, sericite, and topaz in nearbyoutcrops of Flying W basaltic lava breccia, and veinletsof tourmaline, qvartz, and beryl along fractures in FlyingW conglomerate outcrops. These last veinlets appearidentical to the beryl-bearing veinlets in the Dysart area.

In the Flying W rhyolite, the altered and veined rockis bleached to a pink or white color, with pervasive mi-croveinlets containing quarlz, topaz,sericite, and fluorite.In thin section, these veinlets contain topaz and sericitealong the margins, with quartz and minor fluorite in thecenter. The pink to white, bleached areas surroundingthese microveinlets contain primary plagioclase gtains al-tered to sericite plus topaz.

DrscussroN

Origin of topaz in Flying W rhyolite

In the Spring Creek area, topaz is found only in visiblyaltered samples, either as an alteration product ofplagio-

Fig. 10. Photograph of blue beryl on tourmaline-fluorite-quartz vein, Dysart area.

clase or in veinlets of qtartz, white mica, and topaz (acommon greisen assemblage). In thin section, 15 rela-tively fresh, unaltered, dense, dark red Flying W rhyolitesamples contained no topaz. These facts indicate that thetopaz in the Flying W rhyolite formed as a secondaryhydrothermal mineral. There is no evidence that the to-paz formed as the product of vapor-phase crystallizationfrom a F-rich lava (Kortemeier and Burt, 1986). TheFlying W rhyolite is, therefore, not a Precambrian ana-logue ofTertiary topaz rhyolites (as hypothesized by Burtet al., 1982).

Origin of ongonite and topazite dikes

The Breadpan dike was first described (Meeves, 1966)as a "fluorite-bearing sandstone." This designation mayhave been based on the rock's fine grain size, abundanceof quartz, and friable weathered appearance. It has alsobeen suggested to us verbally that these dikes representgreisen veins or greisen replacement of more normalgranitic dikes. Nevertheless, the evidence supporting aprimary magmatic origin is substantial and includes (1)crosscutting relationships with country rocks, (2) chilledmargins along contacts with country rocks, (3) flow folia-tion along dike margins, (4) flow lineations on dike mar-gins, (5) inclusions of country rock (entrained xenoliths)along dike margins, and (6) abundant topaz without re-placement textures and with an igneous g'eochemical sig-nature. These features are discussed individually below.

Crosscutting relationships with country rocks. Both theBreadpan and Dysart dikes show definite crosscutting re-lationships with the rocks they intrude. The Dysart dikecuts across the contact between the Flying W conglom-erate and the Flying W basalt. The Breadpan dike, ori-ented approximately parallel to the surrounding Bread-pan Formation, locally contorts and cuts across thefoliation.

Chilled margins. Margins of the Breadpan and Dysartdikes exhibit a much finer grain size (0.01 mm) as com-pared with the interiors (grains up to I mm). The chillingefect of the country rock on the dikes can be observedin the space of a single thin section taken at the dikFcountry rock contact (Fig. I l). Grain size doubles fromthe contact to a point less than 1 cm inside the dike.

KORTEMEIER AND BURT: ONGONITE AND TOPAZITE DIKES

Fig. 11. Photomicrograph of chilled margin, Breadpan dike. Darker, higher-relief grains are topaz. Remaining grains are quartz.Plane-polarized light. Field ofview about 700 pm across.

5 1 5

Flow foliation. The margins of all dike bodies exhibita pronounced flow-foliated texture. In hand specimen,the foliated appearance is due to pink and white color-banding parallel to the contact of the dike and countryrock (Fig. 4). This banding occurs both as linear zones,or as highly contorted zones (Fig. l2). In thin section, itis evident that the foliation is caused by segregation ofquartz (white) and topaz (pink) into distinct zones parallelto the margins (Fig. l3). The prismatic topaz grains inthese zones are oriented parallel to the margins of therock and to the flow foliation; they are not oriented per-pendicular to the dike-country rock contact as would beexpected if this color and mineral zoning representedbanding in a fracture-filling hydrothermal vein. This par-allel orientation would likewise not be expected of sec-ondary topaz replacing other minerals.

Flow lineations. Lineations observed on the flow-foli-ated margins of the Breadpan and Dysart dikes (BreadpanIineations: plunge 54", bearing N6'E, Dysart lineations:39', N68'E) indicate intrusion of a viscous fluid (magma)and indicate that this magma was intruded upward ratherthan horizontally. These lineations therefore suggest thatthe igneous body from which the dike magma emanatedlies beneath the study area rather than some distance awayfrom it laterally.

Inclusions of country rock (xenoliths). The Breadpandike, the Dysart dike, and the dikelet east ofSpring Creekall contain entrained pieces of country rock along theirmargins (see Fig. 8). The inclusions are completely sup-ported by the enveloping dike rock. Flow foliation pres-ent at the margins of the dikes is distorted around these

xenoliths. A thin section of an inclusion from the Bread-pan dike shows alteration ofthe xenolith to sericite.

Topaz without replacement textures. The majority ofthe topaz in the dikes ofthe Flying W ranch area occursas small, acicular grains (Fig. 5) or as elongate, radiatinggrains (Fig. 6). This is in contrast to topaz textures inhydrothermally altered rocks such as the Mount BischoflTasmania and East Kemptville, Nova Scotia tin greisens(Groves et al., 1972; Richardson, 1985; Chatterjee, et al.,1985), the Henderson porphyry molybdenum deposit,Colorado (Wallace et al., 1978; Seedorfl 1985), and theMount Pleasant porphyry tin-tungsten deposit, NewBrunswick (Pouliot er al., 1978; Parrish, 1982). Topaz inthese rocks occurs as masses ofanhedral grains often re-placing other phases. Acicular groundmass topaz, whichalso occurs in igneous ongonite dikes from Mongolia (Ko-valenko and Kovalenko,1976; see discussion below) ap-pears to be the product of magmatic rather than hydro-thermal processes.

Topaz with an igneous geochemical signature. Until re-cently, topaz was generally accepted to be primarily theproduct of postmagmatic pneumatolytic processes suchas greisenization or vapor-phase crystallization (Eading-ton and Nashar, 1978; Clarke, l98l). Fluid-inclusion andexperimental studies (Kovalenko and Kovalenko, 1976;Eadington and Nashar, I 978) indicate that topaz can alsoform under magmatic conditions as a liquidus phase (seethe ongonite and topazite discussions below). The occur-rence of abundant topaz in these dikes, therefore, doesnot preclude a magmatic origin.

In an ion-microprobe study of topaz from different en-


Fig. 12. Photograph of contorted flow foliation, Breadpandike.

vironments, Hervig et al. (1987) discovered that concen-trations ofLi and B in topaz vary considerably dependingon the geologic environment in which the topaz grew.Topaz samples from six topaz rhyolites, five pegmatites,and four hydrothermal greisens were analyzed. (Singlesamples of topaz from a metamorphic rock, an xenolithof granite in rhyolite, and a porphyry molybdenum de-posit were also analyzed and are included in Fig. 14, butwill not be discussed here.) Topaz crystals from rhyolitesare about 100 times richer in Li than are those fromhydrothermal greisens or pegmatitic rocks. B is also high-est in rhyolitic topaz, but the variation between the dif-ferent geologic environments is less pronounced than thatcreated by the Li values.

It is assumed that the variations in bulk compositionof the various rock types are not responsible for the Liand B results: pegmatites are enriched in B and Li, yettopaz from pegmatites shows low concentrations of bothelements. Hervig et al. (1987) proposed that variations intemperature, pressure, cooling rate, and growth mediummay cause the variations in Li and B concentrations. Rel-atively high-temperature, low-pressure, rapidly cooledmagmatic rocks such as rhyolites tend to have greatersubstitution of B and Li in topaz than more slowly cooled,low-temperature, high-pressure, aqueous-phase-presentrocks such as pegmatites and greisen veins and replace-ments.

Ion-microprobe analyses of topaz free of replacementtextures from the Dysart ongonite dike, Flying W rancharea, show striking similarities to rhyolitic topaz analyses,with relatively high concentrations of both Li and B. Whenplotted on a Li-B concentration diagram with rhyolitic,hydrothermal, and pegmatilic topaz, topaz from the Dy-sart ongonite falls well within the rhyolitic topaz field(analysis 16, Fig. l4).Topaz from a specimen of Bread-pan dike topazite (analysis 17, Fig. l4) is somewhat lowerin Li and B, but still lies well outside the greisen andpegmatite fields of other analyses. This evidence suggests

thattopaz in the dike rocks from the Flying W ranch areawas quenched from magmatic conditions, rather than re-crystallized under lower-temperature hydrothermal con-ditions.

The features described above indicate that the dikeswere emplaced and crystallized as magmatic bodies.Nevertheless, some areas of these bodies have evidentlyundergone syn- or postmagmatic alteration. Quartz vein-lets, fluorite "blebs," and zones altered to sericite, qvarrz,fluorite, and lopaz (where the topaz occurs as masses ofanhedral grains), all suggest that the dikes have locallybeen affected by hydrothermal fluids. These fluids pre-sumably were released late in the crystallization historyof the dikes or underlying intrusive body (Kovalenko andKovalenko, 1976, 1984). Intrusion of the dike-formingmagma along steeply dipping planes (as indicated by thepresent dike orientations and the presence ofnear-verti-cal flow lineations along the margins of the dikes) couldhave been followed by "streaming" of late-stage fluids upthe dike, causing the observed alteration.

Relationship of mineralization to dike emplacement

The three mineralized zones studied in the Flying Wranch area (the Breadpan, Dysart, and Spring Creek) sharea common alteration assemblage that consists predomi-nantly of quartz, tourmaline, fluorite, sericite, beryl, andtopaz. Differences among the areas appear to be due tovariations in the physical and chemical properties of thecountry rock and of the mineralizing fluids. The two mostextensive areas of alteration and mineralization (Bread-pan and Dysart areas) are both close to ongonite or to-pazite dikes, whereas the area of minor alteration andmineralization (Spring Creek) lies 600 m from an ob-served dike (see Fig. 2). On the basis of these field obser-vations, mineralization and dike emplacement wereclosely associated, and the extent of mineralization is re-lated to distance from a dike. Fluids given offby the dikesand their parent igneous body evidently altered and min-eralized the surrounding (or nearby) country rock. Morefluids than those given off by the dikes themselves areneeded to account for the two main areas of mineraliza-tion and also to account for the occulTence of weak al-teration throughout the study area, including the east sideof Spring Creek.

Age of dike emplacement and associated mineralization

Given that the emplacement of the ongonite and to-pazite dikes was related to the mineralization, the age ofthese two events should be comparable. Both the dikesand the mineralization affect Breadpan Formation andFlying W Formation rocks, dated at l7I5 Ma (Conway,1976). The dikes also cut across the foliation of theserocks and therefore postdate the Mazatzal orogeny, esti-mated to have occurred in the interval 1660-1715 Ma(Silver, 1964). These data indicate that the maximum ageof the dikes and associated mineralization is about l7l5Ma.

A minimum age for the intrusive and mineralizingac-


Fig. 13. Photomicrographof flow-foliatedmarginof Dysartdike.Noteabundance of topazinzoneonleft,abundanceofquartzon right. The segregation of topaz and quartz into discrete zones parallel to dike maryins gives the rock a banded appearance inhand specimen. Plane-polarized light. Field ofview about 700 pm across.

5r7

tivity is more difficult to determine. Apache Group rocks,deposited within the interval 1500 to ll00 Ma (Living-ston and Damon, 1968; Silver et al., 1977 a, 1 977b; Shaf-iqullah et al., 1980) are present at several locations in thefield area and at one location lie not more than 250 mfrom Dysart dike outcrops (see Fig. 2). The dike, how-

ever, pinches out before reaching the contact with theyounger Precambrian rocks, and therefore its relative ageis not indicated. A similar problem exists at the western-most end of the Breadpan dike, where Apache Grouprocks crop out approximately 400 m west of the last out-crop ofthe dike.

-t_

l og L i (ppmw)

Fig. 14. Li vs. B diagram from ion-microprobe analyses of topaz from different geologic environments (from Hervig et al.,1987). The Dysart dike ongonite sample is area 16, the Breadpan dike topazite is area 17, and the topazite from N.S.W., Australia,is area 18.

I tt ,Q 1

g i 4

@


The mineralization may be more revealing. In thenortheastern part of the map area (Fig.2), Breadpan andFlying W Formation rocks are veined and altered in thesame manner as rocks described from the Spring Creeklocality. Overlying these rocks and separated from themby an angular unconformity is a conglomerate of theApache Group. Veined and altered Alder Group rocksoccur immediately below the unconformity, but no vein-ing or alteration is apparent in the conglomerate above.This evidence, combined with the lack of dikes cuttingthe Apache Group rocks, suggests that the age ofthe on-gonite and topazite dikes and associated mineralizationis greater than that of the Apache Group rocks.

From the above arguments, the age of the dikes andmineralization probably falls within the interval 1715 toI100 Ma. This activity may have been contemporaneouswith the intrusion of 1410-1460 Ma anorogenic, F-richgranites (Silver, 1968; Silver et al., 1977a. 1977b) suchas the Ruin and Lawler Peak intrusions. Nevertheless, ayounger (e.g., Tertiary) age cannot definitely be excluded.

Comparison of Flying W dikes to other rock types

The dikes in the Flying W ranch area represent typesofigneous rocks that have not previously been describedin North America. They have some of the textural fea-tures commonly attributed to fine-grained aplite dikes,yet the dikes of the Flying W ranch area are not trulyaplitic, or allotriomorphic-granular in texture. A trueaplitic texture consists ofequal-sized grains, with no grainsbetter formed than others (Shand, 1969, p. 185). Rocksfrom the Flying W area dikes are better described as fine-grained porphyritic; distinctions in grain size betweengroundmass and phenocrystic phases are easily made, andthe habit of the individual minerals varies from euhedralto anhedral.

The dikes ofthe Flying W ranch area also differ fromtypical aplitic dikes by their lack of associated pegmatiticfabrics. Pegmatites and aplites are believed to be gener-ated in the same geologic environment and are often in-timately associated (Hatch et al., 1972, p. 222; Williamset al., 1982, p. 178). The lack of associated pegmatiticfabric in the Flying W area dikes indicates either that (l)the dike-forming magma was relatively "dry" (did notcontain an immiscible fluid phase) at the time of em-placement or (2) a high level of intrusion and subsequent-ly low lithostatic pressures led to the loss of any immis-cible fluid phase at the time of dike emplacement (or thedikes simply cooled very rapidly). In any case, the FlyingW dikes do not represent typical dikes of aplite-pegmatiteaffinity.

Mineralogically, dikes of the Flying W ranch area aresimilar to tin-bearing greisens such as occur in Cornwall,England (Exley and Stone, 1964; Hosking, 1964). Grei-sens are formally defined as pneumatolytically alteredgranite rocks (or surrounding country rocks) composedlargely of quartz, mica, tourmaline, and topaz (Johann-sen, 1932, p. 5; Will iams et al., 1982,p. 172; Bates andJackson, 1987, p. 290). Convincing evidence ofreplace-

ment is usually present, such as aggregates of white micaafter feldspar (Hatch et al., 1972, p. 223), partial rims oftopaz around quartz phenocrysts, or topaz pseudomorphsafter feldspar (Groves et al., 1972, p. 195-197)' In con-trast to the situation in the Flying W area, greisens arecommonly associated with and localized along sets of hy-drothermal veins.

The topazite and ongonite dikes share some importantsimilarities with topaz rhyolites: both rock types are be-lieved to be produced by extreme differentiation of gra-nitic magma, and both rock types may be associated withberyllium-fluorine mineralization (Bikun, 1980; Burt etal., 1982 Burt and Sheridan, 1986). Nevertheless, thedikes of the Flying W ranch area appear to be more highlydifferentiated than topaz rhyolites (their relatively highSr content may be due to contamination by wall rocks,as discussed below).

Two rock types strikingly similar to those of the FlyingW ranch area are ongonites, described from Mongoliaand the Transbaikal region of the USSR (Kovalenko etal., l97l; Kovalenko et al., 1975; Kovalenko and Ko-valenko, 1976), and topazites (sometimes called silex-ites), described from the New England district, New SouthWales, Australia (Eadington and Nashar, 1978; Eading-ton, 1983; Kleeman, 1985). These two rock types are de-scribed below.

Ongonites. Ongonites are defined (Kovalenko et al.,l97l) as the aphanitic, vitrophyric, or phenocrystic sub-volcanic analogue of rare-metal Li-F granites. Kovalenkoetal. (1971,1975) and Kovalenko and Kovalenko (1976)describe ongonites in Mongolia and the Transbaikal re-gion ofthe USSR as topaz-beaing"quartz keratophyres"that form subvolcanic dikes and stocks, as well as lavaflows. Ongonite dikes from the type locality, the Ongon-Heierhan tungsten belt in central Mongolia, are describedbelow in a condensed extract from an unpublished En-glish translation of Kovalenko and Kovalenko (1976).

Ongonite dikes occur 100 km southwest of Ulan Bator, Mon-golia, in the Ongon-Heierhan tungsten belt approximately l0 kmeaSt of exposures of the Mesozoic Ongon-Heierhan granite mas-sif. The ongonite dikes pinch out near the Ongon-Heierhan gran-ite outcrops, and no relationship between the two has been dis-cerned. Gravimetric evidence, however, places granites ofOngon-Heierhan type at a depth of approximately 600 m underthe ongonite dike system. K-Ar dates of micas from both theOngon-Heierhan massif and the ongonite dikes are early Me-sozorc.

The ongonite dikes, ranging in width from less than 20 cm to2.5 m and in length from less than 100 to 450 m, cut fine-grained, weakly metamorphosed, mid-Paleozoic sandstones ofthe Khent series. Xenoliths of these sandstones are commonalong the margins of the dikes. The dikes' contacts \,\rith thewallrock and with xenoliths are extremely sharp, and neither thewallrock nor the xenoliths show signs of alteration or contactmetamorphism.

In outcrop, the dike rock is white and dense, with conchoidalfracture. Veinlets ofquartz and sericite with tungsten and tin oreminerals cut the dikes. The texture of the dikes ranges fromaphyric or glassy in a marginal chill zone 10 to 15 cm wide toporphyritic and phenocryst-rich in the centers. In general, thecrystallinity tends to increase toward the center of the dikes. The

KORTEMEIER AND BURT: ONGONITE AND TOPAZITE DIKES 5 1 9

aphyric marginal zones are commonly flow-banded parallel tothe contact. Contortion ofthe flow-banding occurs around xeno-liths.

The primary minerals of ongonites are albite, potassium feld-spar, and quartz. These three minerals occur both as phenocrystsand in the microcrystalline groundmass. Topaz and micas areless abundant and occur only rarely as phenocrysts. Accessoryminerals include fluorite, garnet, zircon, ilmenite, columbite,tantalite, cassiterite, and pyrite. Needle-shaped, microlitic topazcrystals, less than 0.05 mm long and 0.01 mm wide, may con-stitute up to 100/o of the groundmass of ongonites and are con-sidered to be a characteristic feature ofthis rock. Along the aphyricflow-banded margins of the ongonite dikes, topaz microlites areoriented subparallel to the contacts ofthe dikes with the countryrock and appear to "flow around" phenocrysts. These featuresindicate that the topaz microlites existed, if only in part, whilethe ongonitic magma was still molten and mobile. Segregationof these topaz microlites into discrete zones (together with ori-ented albite phenocrysts and alternating glassy and crystallizedzones) produces the flow-banded appearance ofthe marginal areasof the dikes.

Postmagmatic alteration was extremely minor and did notsubstantially influence either the composition or the fabric oftherock.

The major-element geochemistry of a dike from the Ongon-Heierhan district, averaged from 53 analyses, is given in Table1 (this paper). It ditrers markedly from that of all other magmaticrocks. Although the silica content resembles that ofgranites andrhyolites, the alumina value approaches that of a syenite. Thehigh alumina is due to the significant amounts of topaz in therock. Ongonites are also very leucocratic and contain almost noTi (similar to alaskite). The ratio K/Na is almost always greaterthan 1.

The rare-element geochemistry of the ongonite dike, above, isalso presented in Table 1. Relative to "average concentrationsin granites," ongonites are enriched in F, Li, Rb, Tl, Be, Sn, Hf,Nb, and Ta. Concentrations of Ba, Sr, REEs, Y , and Zr are low.Low values for LiiRb, Nb/Ta, andZr/Hf are also characteristic.Overall, the rare-element geochemistry of these rocks is similarto that of rare-metal pegmatites and rare-metal Li-F granites.

Kovalenko and Kovalenko (1976) indicate that all of

their geologic, petrologic, and geochemical data support

the genesis of ongonites by magmatic processes. They

considered the Ongon-Heierhan ongonite bodies to be

subvolcanic and rapidly cooled, as evidenced by por-

phyritic textures, glassy to microcrystalline groundmass,

and other features "ofeffusive appearance" in the ongon-

ite bodies.Kovalenko and Kovalenko (197 6, I 984) stated that on-

gonites may be regarded as residual melts derived from

the protracted fractional crystallization of crustally de-rived F-rich granitic magmas, possibly of alaskitic com-position. The elevated F contents of this granitic magma

would lead to Na and Al enrichment in the remaining

melt (Manning, l98l; Manning et al., 1981; Pichavant

and Manning,1984; Dingwell, 1985). This F-, Na-, and

Al-enriched melt from which ongonites crystallize should

correspond to one of the most strongly differentiatedgranitic magmas possible in nature. Furthermore, its vis-

cosity is much lower than that of a normal granitic melt(Dingwell, 1987), facilitating its intrusion as narrow dikes.

Experiments performed on the system ongonite-Hro-HF (Kovalenko, 1979) demonstrate the feasibility of

crystallizing the major minerals of ongonites (albite, mi-

crocline, qvartz, topaz, atd mica) from a silicate melt.Crystallization of ongonitic melts is completed at verylow temperatures: 575(25) 'C. If the melt contains B aswell as F (as suggested by the abundant tourmaline in theveinlets near the dikes of the Flying W ranch area), its

crystallization may be further depressed (to temperaturesless than 500 "C: London, 1986, 1987). During the finalstages of crystallization of the ongonitic magmas, the fluidphase becomes enriched in Be, W, Pb, Zn, F, B, Li, Si'and HrO, components responsible for greisenization. Thepartitioning of these elements into the fluid phase at theend of crystallization helps explain the relative dearth ofthese elements in the ongonite bodies themselves and theabundance of these same elements in nearby rocks (i.e.,

Ongon-Heierhan tungsten seams; Breadpan-Dysart be-

ryllium mineralization).Mineralogically, the main part of the Dysart dike re-

sembles the Ongon-Heierhan ongonites. Both dikes con-sist predominantly of quartz, lopaz, and feldspars. Feld-spar compositions differ, however, with theOngon-Heierhan ongonite containing abundant potas-

sium feldspar and the Dysart ongonite containing only

albitic plagioclase. (K-feldspar in a dikelet on the east sideof Spring Creek is mentioned above.)

Textural and outcrop characteristics of the ongonitedikes from Mongolia are strikingly similar to those of the

dikes in the Flying W ranch area (see Flying W ongoniteand topazite dike descriptions above). The multiple sim-ilarities between the Ongon-Heierhan dikes and the dikesofthe Flying W ranch area suggest that they shared com-parable magmatic, intrusive, and cooling histories. Dif-ferences between the dike systems in the two localities(i.e., higher Sr and Ba, lower Rb in the Flying W dikescompared to lower Sr and Ba, higher Rb in the Mongo-Iian dikes) could result from differences in the initialcomposition of the two source magmas or differences infractionation trends between the two magma systems, aswell as differences in final mineralogy. The higher Sr and

Ba in the narrow Flying W dikes could also result fromcontamination by metavolcanic country rocks.

Topazites. Quartz-topaz rocks from the New Englanddistrict, New South Wales, Australia, occur as dikes and

sills that intrude a biotite granite and siltstones of a roofpendant in the granite. Eadington and Nashar (1978) re-ferred to these intrusives as topazites and indicated thatthese rocks have mineralogic, textural, and field relation-ships that suggest a magmatic origin. This contention wasdisputed by Kleeman (1985), who concluded lhat thequartz-topaz assemblage formed by hydrothermal re-placement of late microgranite dikes and sills (i.e., bypervasive greisenization), with magmatic grain size andtextures preserved.

The thickness of the topazite intrusive bodies rangesfrom a few centimeters to tens of meters. Boundaries withthe enclosing rocks are sharp (Eadington, 1983), and where


the topazite intrudes siltstones, a narrow zone of indura-tion, silicification, or metamorphism to biotite-quartzhornfels is developed, the degree ofalteration dependingon the size of the intrusion.

In hand specimen, the New South Wales quarlz-topazrock resembles a quartzite, but contains 5 vo10/o miaroliticcavities. Quartz and topaz are the only major minerals;wolframite, muscovite, and tourmaline are minor. Eu-hedral to subhedral topaz crystals constitute l8-27o/o ofthe rock; the remainder is mainly qvarlz, which is typi-cally anhedral.

Chemical analyses of the NSW topazite (Table l) showextremely low alkali-metal and alkali-earth contents andhigh SiOr, AlrOr, and F contents, as would be expectedfrom the mineralogy.

Locally, hydrothermal alteration is believed to haveafected the topazite bodies (Eadington and Nashar, I 978).In places, topaz has been replaced by white mica, andquartz grains have been recrystallized to produce a com-pact rock consisting primarily of quartz and mica. Thisaltered rock often contains economic concentrations ofwolframite (Eadington, 1 983).

The granite into which the topazite bodies intrude isreferred to as the Mole Granite and is one of three graniteplutons within the New England batholith that host tinmineralization (Eadington, 1983). Chemical analyses ofthe granite show high silica values (74.1-77.3 wto/o) andhigh concentration of the trace elements F (0.30lo) and Li(80-620 ppm). In view of the intrusive relationship of thetopazite into the Mole Granite and the high F contentsofthe granite, Eadington (1983) st'ggested that the topa-zite bodies developed as late-stage, magmatic derivativesof this granite. Partitioning of Na and K into the aqueousphase (possibly owing to high halogen concentrations:Eadington and Nashar, 1978) would leave a silicate meltenriched in Al and F that would then crystallize quartzand topaz. Studies of primary fluid inclusions in topazfrom the topazite (Eadington and Nashar, 1978) indicatethat crystallization of topazite from a silicate melt couldoccur in the temperature range of 850 to 570 "C followingsaturation of a granitic magma with water (second boil-ing) and the formation of immiscible silicate and aqueousphases. In topazites from the New England district, mia-rolitic textures and high porosities are evidence of an im-miscible system; the cavities are thought to have beenoccupied by the aqueous phase at the time of crystalli-zalror.

Processes similar to those proposed for the formationof New England district topazites may be responsible forthe exotic mineralogy and geochemistry of the Breadpantopazite dike in the Flying W ranch area. Composition-ally, at least, these two topazite occurrences are nearlyidentical (Table l). Although textural evidence of an im-miscible aqueous phase in the Breadpan topazite is scarce(miarolitic cavities are few, and the porosity of the rockis low), the F-, Be-, K-, and Na-rich mineralized area inclose proximity to the topazite dike suggests that anaqueous phase developed during the crystallization of the

dike or its parent pluton and was lost to the surroundingcountry rock.

The extremely low Li and B contents in topaz fromNew South Wales topazite (analysis 18, Fig. 14) presum-ably imply that it was generated by an aqueous, ratherthan magmatic, fluid, at least for this sample (which rep-resents a single coarse-grained specimen of unknown set-ting that was donated to D.M.B. by a visiting miningcompany geologist). This evidence, although it is incon-clusive, supports the greisenization hypothesis of Klee-man (1985). The higher Li and B contents and distinctivetextural features of topaz from the Breadpan dike topa-zite, on the other hand, appear to support a magmaticorigin for this rock.

Relationship of topazite to ongonite

In neither the New England district nor the Ongon-Heierhan district do ongonite and topazite occur togeth-er. The occuffence ofboth ongonites and topazites in theFlying W ranch area therefore provides a unique oppor-tunity to examine the relationship between these two rocktypes.

The transition from one rock type to the other has beenobserved at two localities. At the northern end of theDysart dike, ongonite gives way to topazite; at the west-ern end of the Breadpan dike, topazite grades into on-gonite. Both transitions occur over less than 5 m. Theprocesses producing this gradation from one rock type toanother are not well understood, and information ob-tained from thin-section examination is inconclusive.

Observations of anhedral, partly resorbed albite phe-nocrysts in the ongonite of the Dysart dike suggests thatearly-formed albite grains were not in equilibrium withthe final melt. In the feldspar-free topazite of the Dysartarea, total resorption of plagioclase grains may have re-sulted in the partitioning of Na into an immiscible fluidphase, leaving an alkali-depleted melt to crystallize topazand quartz. Similarly, ongonitic rock of the predomi-nantly topazitic Breadpan dike may represent areas whereequilibrium between plagioclase and melt existed to thefinal stages of crystallization, because the system was moreclosed to loss ofaqueous fluids. In thin section, ongoniticrock from the Breadpan dike exhibits a fine-grained, al-lotriomorphic-granular texture, with interlocking quartz,plagioclase, and topaz grains. This is the only area in thedike where this equilibrium texture is so well developed.

In general, topazitic rocks have lost all of their alkalicomponents, probably during crystallization, whereas on-gonitic rocks have retained a large portion of their orig-inal Na.

Experiments on the system ongonite-HrO-HF by Ko-valenko and Kovalenko (1976) and Kovalenko (1979)and on the system granite-Hro-HF by Glyuk and Anfi-logov (1973) and Kovalenko (1978) indicate that F con-centrations strongly influence the phases that crystallize.Kovalenko and Kovalenko's (l 976) reconnaissance phasediagram of the system ongonite-HrO-HF is shown in Fig-ure 15. At low HF concentrations, the typical minerals


of ongonites coexist: albite, potassium feldspar, quartz, T'Ctopaz, and mica (fields I and II). At high HF concentra- 800tions, the topazite assemblage quafiz + topaz crystallizesexclusively (fields X and XI), as alkalis are partitionedinto the aqueous fluid.

521

' X I l '

These experiments suggest that the changes in the min-eralogy of the ongonite and topazite dikes in the FlyingW ranch area, in particular, the presence or absence offeldspar, may be a function of variations in HF concen-trations within the dikes and in timing and degree of lossofvolatiles to the country rocks.

CoNcr-usroNs

Field, textural, geochemical, and mineralogical char-acteristics ofthe topaz-bearing dikes in the Flying W rancharea of the Tonto basin all support the genesis of thesebodies from a magma. Several lines of evidence contra-dict an origin for these bodies by greisenization or vein-mg.

The ongonite and topazite dikes ofthe Flying W rancharea are texturally, and, in part, mineralogically, similarto magmatically derived, subvolcanic ongonite dikes lo-cated in the Ongon-Heierhan district, Mongolia, and mayshare a similar magmatic, emplacement, and cooling his-tory with these Mongolian dikes. The Dysart dike, con-sisting predominantly of quartz, albite, and topaz, is anongonite. The Breadpan dike, lacking albite and consist-ing predominantly of quartz andtopaz, is mineralogicallyand compositionally similar to topazite dike rocks de-scribed from the New England district, New South Wales,Australia. This occurrence ofongonite and topazite is thefirst of its kind in the United States and is the only rec-ognized location in the world where the two rock typesoccur together.

The ongonite- and topazite-producing melts are be-lieved to be the product of extreme fractional crystalli-zation of an F-rich granitic source magma. Elevated Fconcentrations in the fractionating source magma leadsto Na and Al enrichment in the remaining melt. Afterdike emplacement, this melt would crystallize the ongon-itic phases quartz, albite, and topaz (equivalent to theDysart dike). Extremely high HF concentrations in themagma are needed to produce topazitic compositions, forwhich feldspars do not crystallize, leaving quartz and to-paz as the only major phases. Observed transitions fromongonite to topazite, and vice versa, may be the result offluctuating HF concentrations within the dikes and ofvariations in volatile loss by the dikes as they crystallized.

The lack of coexisting pegmatitic fabrics in the dikesof the Flying W ranch area indicates that the intrudingmagma was originally "dry" (i.e., no immiscible fluidphase was present at the time of emplacement of thedikes). During the final stages of crystallization of the dikebodies, however, an immiscible fluid phase was lost tothe surrounding Alder Group country rocks. The expul-sion of this fluid phase, enriched in K, Na, F, B, and Be,contributed to the alteration and mineralization at theBreadpan, Dysart, and Spring Creek localities. The

VII

V I I I

0.1 0.5 1.0 r.3 3.4 3.8

Wt. % HF in system

Fig. 15. Phase diagram for the systern ongonite-HrO-HF at1000-atm pressure and fluid: rock ratio of 5:1. Dots representanalytical data points. Fields are as follows:

I . A b + K F + Q z + M i + F lII. Ab + KF + Top + Qz + Mi + M + Fl

I I I . A b + I ( F + Q z + M + F lI V . A b + M + F IV . Q z + M + F l

V I . A b + Q z + M + F lV I I . Q z + T o p + M + F l

V I I I . Q z + T o p + A b + M + F lI X . M + F IX . Q z + T o p + F l

X I . Q z + T o p + M l + M 2 + F lXII. MI + M2 + FI

Abbreviations: Ab : albite; KF : potassium feldspar; Qz :quartz; Top : topaz; Mi : mica; M, Ml, M2 : melt(s); Fl:fluid (from Kovalenko and Kovalenko, 1976).

strongest alteration and mineralization occur in the Dy-sart and Breadpan areas, where mineralized areas anddikes are in close proximity. The alteration assemblageassociated with dike emplacement includes tourmaline,quartz, fluorite, potassium feldspar, calcite, beryl, and to-paz in approximate decreasing order of relative abun-dance.

Topaz in the Flying W rhyolite tuff on the west side ofSpring Creek was formed as part of the alteration andmineralization associated with dike intrusion. Topaz inthe rhyolite occurs as part ofa typical greisen assemblagein veinlets in the rock and as an alteration product ofplagioclase. There is no evidence that topz in this unitformed during vapor-phase crystallization in the rhyolite,as is the case for topaz in Tertiary topaz rhyolites. Inother words, the Flying W is not a Precambrian exampleof a topaz rhyolite.

Dike emplacement and associated mineralization ap-pear to be post-Mazatzal orogeny and pre-Apache Groupin age. This would imply that both dikes and mineraliza-tion were formed during the interval 1715-1100 Ma.

Acxxowr,nlcMENTS

We would like to thank S. B. Keith of MagmaChem, Inc., and theGraduate Student Association ofASU for partial financial support ofthis

700

' X I '

l a a/ X

a a

522

project R L. Hervig of ASU conducted the ion probe study of topaz fromdifferent environments and is thanked for including Flying W dike rocksin that study. The authors also acknowledge the excellent work previouslycarried out in the region by R. G. Gastil, now of University of Califomia-Santa Barbara, and C M. Conway, now with the U.S. Geological Survey,Flagstaff. Finally, we thank V. I. Kovalenko of the IGEM, USSR Acad-emy of Sciences, Moscow for his field and laboratory insights relating tothis area during a US-USSR Academy of Sciences exchange visit in thefall of 1986, and D London and P. Cernj'for careful reviews ofan earlierversion of this paper Revision of the manuscripl was performed whilethe second author was a Visiting Scientist at the Lunar and PlanetaryInstitute, Houston, which is operated by the Universities Space ResearchAssociation under Contract No. NASW-4066 with the National Aero-nautics and Space Administration. This paper is Lunar and PlanetaryInstitute Contribution no. 648

RnrsnnNcrs cITED

Bates, R.L., and Jackson, J.A., Eds (1987) Glossary ofgeology (3rd edi-tion) American Geological Institute, Alexandria, Virginia.

Bikun, J.V (l 980) Fluorine and lithophile element mineralization at SporMountain, Utah, M.S thesis, Arizona State University, Tempe.

Burt, D M., and Sheridan, M F (l 986) Mineral deposits related to topazrhyolites in the southwest. Arizona Geological Society Digesr,16, 170-I 78.

Burt, D M , Sheridan, M.F., Bikun, J V., and Christiansen, E.H. (1982)Topaz rhyolites-Distribution, origin, and significance for exploration.Economic Geology, 77, 1818-1836.

Chattedee, A.K., Strong, D.F., and Clarke, D.B. (1985) Petrology of thepolymetallic quartz-topaz greisen at East Kemptville In A.K. Chatter-jee and D.B. Clarke, Eds., Guide to granites and mineral deposits ofsouthwestern Nova Scotia, p- 152-196. Nova Scotia Department ofMines and Energy, Halifax

Christiansen, E.H., Burt, D.M., Sheridan, M.F., and Wilson, R.T. (1983)The petrogenesis of topaz rhyolites from the western United States.Contributions to Mineralogy and Petrology, 83, l6-30.

Christiansen, E H., Sheridan, M F., and Burt, D.M. (1986) The geologyand geochemistry ofCenozoic topaz rhyolites from the western UnitedStates Geological Society ofAmerica Special Paper 205.

Clarke, D.B. (1981) The mineralogy of peraluminous granites: A review.Canadian Mineralogist, 19, 3-17 .

Conway, C.M. (1976) Petrology, structure, and evolution ofa Precam-brian volcanic and plutonic complex, Tonto basin, Gila County, Ari-zona. Ph.D. thesis, California Institute ofTechnology, Pasadena.

Conway, C.M., McColly, R.A., Marsh, S.P, Kulik, D M., Martin, R.A.,and Kilburn, J.E. (1983) Mineral resource potential of the Hells GateRoadless Area, Gila County, Arizona U.S Geological Survey Sum-mary Report (to accompany Map MF-1644-A).

Dingwell, D B. (1985) The structure and properties offluorine-rich silicatemelts: Implications for granite petrogenesis (extended abs.). In R.P.Taylor and D.F. Strong, Eds., Granite-related mineral deposits. Geol-ogy, paragenesis, and tectonic setting, p. 72-81. Canadian Institute ofMining and Metallurgy, Geology Division.

-(1987) Melt viscosities in the system NaAlSirO'-HrO-FrO ,. InB.O. Mysen, Ed., Magmatic processes: Physicochemical principles, p.423431. Geochemical Society Special Publication No I

Eadington, P.J (1983) A fluid inclusion investigation ofore formation ina tin-mineralized granite, New England, New South Wales. EconomicGeology, 78,1204-1221

Eadington, P.J., and Nashar, Beryl (1978) Evidence for the magmaticorigin ofquartz-topaz rocks from the New England batholith, Austra-lia. Contributions to Mineralogy and Petrology, 67, 433438.

Exley, C.S., and Stone, M. (1964) The granite rocks ofsouthwest England.In K.F.G. Hosking and G.J. Shrimpton, Eds., Present views of someaspects of the geology of Cornwall and Devon, p. 1 3 l-l 8 L Royal Geo-logical Society of Cornwall, Cornwall

Gastil, Gordon. (1954) Geology ofthe eastern halfofthe Diamond Buttequadrangle, Gila County, Arizona. Ph D thesis, University ofCalifor-nia, Berkeley.

-(1958) Older Precambrian rocks of the Diamond Butte ouadran-


gle, Gila County, Arizona. Geological Society of America Bulletin, 69'

1495-1514.Glyuk, D S., and Anfilogov, V N. (1973) Phase equilibria in the system

granite-H,O-HF at a pressure of 1000 kglcm'z. Geochemistry Interna-

tional, I 0, 32 l-325 (translated from Geokhim iy a, no. 3, 43 U38, I 97 3).

Groves, D.I., Martin, E.L, Murchie, H., and Wellington, H.K. (1972) A

century of tin mining at Mount Bischofl l87l-1971. Tasmanian De-partment of Mines, Geologrcal Survey Bulletin no 54.

Hatch, F.H., Wells, A.K, and Wells, M.K. (1972) Petrology of the igneousrocks. Thomas Murby and Company, london

Hervig, R L., Kortemeier, W T., and Burt, D.M. (1987) Ion-microprobeanalyses of Li and B in topaz from different environments. AmericanMineralogist, 7 2, 392-39 6.

Hosking, K F.G (1964) Permo-Carboniferous and later primary miner-

alization of Cornwall and southwest Devon. In K.F.G. Hosking and

G.J. Shrimpton, Eds., Present views of some aspects of the geology of

Cornwall and Devon, p 201-245. Royal Geologrcal Society of Corn-wall, Comwall

Jackson, K C. (1970) Textbook oflithology. McGraw-Hill, San Francisco.

Johannsen, A. (1932) A descriptive petrography ofthe igneous rocks, vol.

2 University ofChicago Press, Chicago.Kleeman, J.D (1985) Origin of disseminated wolframite-bearing quartz-

topaz rock at Torrington, New South Wales, Australia In High heatproduction (HHP) granites, hydrothermal circulation and ore genesis,

p 197-20l.Institution of Mining and Metallurgy, London.Kortemeier, W.T., and Burt, D.M. (1985) Dikes of ongonite and topazite

from the Tonto basin, central Arizona (abs ). EOS, 66, 1136.-(1986) Origin of topaz in Precambrian metarhyolite from the Flying

W ranch area, Tonto basin, Arizona. Geological Society of AmericaAbstracts with Programs, 18, 367.

Kovalenko, N.I (1978) The reactions between granite and aqueous hy-

drofluoric acid in relation to the origin of fluorine-bearing granites.

Geochemistry Intemational, 14, 108-1 l8 (translated from Geokhimi-ya, no. 4, 503-515, 1978)

-(1919) Experimental investigations of the formation of rare metal

lithium-fluorine granites. Nauka Press, Moscow (in Russian).Kovalenko. V I, and Kovalenko, N.I. (1976) Ongonites (topaz-bearing

quartz keratophyre)-subvolcanic analogue of rare metal Li-F granites.

Nauka Press, Moscow (in Russian)-(1984) Problems of the origin, ore bearing, and evolution of rare

metal ganitoids. Physics ofthe Earth and Planetary Interiors, 35, 5l-

62.Kovalenko, V I., Kuz'min, M.I., Antipin, V S., and Petrov, L.L. (1971)

Topaz-bearing quartz keratoph)Te (ongonite), a new variety of subvol-

canic igneous vein rock. Doklady of the Academy of Sciences of the

USSR, Eanh Science Sections, 199, l3Z-135 (translated from Doklady

Akademii Nauk SSSR, 199, no. 2, 430433, 197 l).Kovalenko, V I., Grebennikov, A.M., and Antipin, V.S. (1975) Ongonite

of the Arybulak stock, Transbaikal: The first find of a subvolcanicanalog ofrare metal-bearing lithium-fluorine granite (apogranite) in the

USSR. Doklady of the Academy of Sciences of the USSR, Earth Sci-

ence Sections, 220, I 58-1 60 (translated from Doklady Akademii Nauk

SSSR,220 , no .5 , l l 69 -1171 , 1975 ) .Livingston, D.E., and Damon, P.E. (1968) The ages of stratifred Precam-

brian rock sequences in central Arizona and northern Sonora Canadian

Journal ofEarth Science, 5, 763-772.London, David. (1986) Magmatic-hydrothermal transition in the Tanco

rare-element pegmatite; evidence from fluid inclusions and phase-equi-

librium experiments American Mineralogist,'7 l, 37 6-39 5.- ( I 987) Internal differentiation of rare-element pegmatites: Effects

ofboron, phosphorus, and fluorine. Geochimica et Cosmochimica Acta,

sr,403420Manning, D.A C (l 98 l) The effect of fluorine on liquidus phase relation-

ships in the system Qz-Ab-Or with excess water at I kb. Contributionsto Mineralogy and Petrology, 76,206-215.

Manning, D A.C., Hamilton, D L, Henderson, C.M.B., and Dempsey,

M.J. (1981) The probable occurrence of interstitial Al in hydrous'

F-bearing and F-free aluminosilicate melts. In C E. Ford, Ed., Progtess

in experimental petrology, 5th Report, 1978-1980, 23-26. NationalEnvironmental Research Council, Publication Series D, 18.

Meeves, H.C. (1966) Non-pegmatitic beryllium occunences in Arizona,Colorado, New Mexico, Utah. and four other states U.S. Bureau ofMines Report oflnvestigations no. 6828

Pamsh, I S. (1982) The Mount Pleasant porphyry tungsten deposit, Can-ada In J Y. Hepworth, Ed., Tungsten geology, Jiangxi, China, p. 165-178 Symposium proceedings, ESCAP RMRDC, Bandung, Indonesia.

Pichavant, M., and Manning, D. (1984) Petrogenesis of tourmaline gran-ites and topaz granites: The contribution ofexperimental data. Physicsofthe Earth and Planetary Interiors, 35, 31-50

Pierce, H.W. (1985) Arizona's backbone: The transition zone. Fieldnotes,Arizona Bureau ofGeology and Mineral Technology, 15, no. 3, 1-3.

Pouliot, Gaston, Barondeau, B, Sauve, P., and Davis, M. (1978) Distri-bution of alteration minerals and metals in the Fire Tower zone atBrunswick tin mine, Ltd , Mount Pleasant area, New Brunswick. Ca-nadian Mineralogist, 16, 223-234

Richardson, J.M (1985) The East Kemptville geisen-hosted tin deposit,Nova Scotia: Overview ofthe geological relationships and geochemis-try (extended abs.) In R.P Taylor and D.F Strong, Eds., Granite-related mineral deposits. Geology, paragenesis, and tectonic setting, p.203-204. Canadian Institute of Mining and Metallurgy, Geology Di-uslon.

Seedorfl Eric. (1985) Cyclic development of hydrothermal mineral as-semblages related to multiple intrusions at the Henderson porphyrymolybdenum deposit, Colorado. (extended abs.) In R P Taylor andD F. Strong, Eds., Granite-related mineral deposits Geology, paragene-sis, and tectonic setting, p. 72-81 Canadian Instjtute of Mining andMetallurgy, Geology Division

Shafiqullah, M., Damon, P.E., Lynch, D.J, Reynolds, S.J, Rehrig, W.A.,and Raymond, R.H. (1980) K-Ar geochronology and geologic historyofsouthwestem Arizona and adjacent areas. Arizona Geological Soci-ety Digest, 12, 201-242-

523

Shand, S.J. (1969) Eruptive rocks. Hafner Publishing Company, New York.Sherlock, S.M. (l 986) Structure and stratigaphy ofearly Proterozoic rocks,

McDonald Mountain-Breadpan Mountain area, Northern Sierra An-chas, Gila County, Arizona. M.S. thesis, Northern Arizona University,Flagstaff.

Silver, L T (l 964) Mazatzal orogeny and tectonic episodicity (abs.). Geo-logical Society ofAmerica Special Paper 82, 185-186.

-(1968) Precambrian batholiths of Arizona (abs.). Geological So-ciety ofAmerica Special Paper l2l, 558-559.

Silver, L.T., Andenon, C.A., Crittendon, M, and Robertson, J.M. (1977a)Chronostratigraphic elements of the Precambrian rocks of the south-western and far western United States Geological Society ofAmericaAbstracts with Programs, 9, 1176

Silver, L.T., Bickford, M.E., and Van Schmus, W.R. (1977b) The 1.4-1.5b y. transcontinental anorogenic plutonic perforation ofNorth Amer-ica. Geological Society of America Abstracts with Programs, 9, 1176-1 1 7 7 .

Smith, L.A. (1962) Mine report-Scorpion beryllium mine. Open-file re-port, Arizona Department ofMineral Resources, Tucson.

Wallace, S.R , MacKenzie, W.B., Blair, R.G., and Muncaster, N.K. (1978)

Geology ofthe Urad and Henderson molybdenite deposits, Clear CreekCounty, Colorado, with a section on a comparison of these depositswith those at Climax, Colorado. Economic Geology, 73,325-368.

Williams, Howel, Tumer, F J., and Gilbert, C.M. (1982) Petrography: Anlntroduction to the study of rocks in thin section (2nd edition). W.H.Freeman and Company, San Francisco

Wilson, E. D. ( I 9 3 9) Precambri an Mazatzal rev olution in central Arizona.Geological Society ofAmerica Bulletin, 50, I I 13-1 164.

MaNuscnrpr RECETvED Jurv 15. 1986M,awuscnrpr AccEprED Jau;.lnv 20, 1988


ongonite and topazite dikes in the flying w ranch area ...see fig. l). beryllium mineralization, in...

Documents